all_results.json

{
    "epoch": 2.999297541394882,
    "eval_logits/chosen": 0.11851656436920166,
    "eval_logits/rejected": 0.21514640748500824,
    "eval_logps/chosen": -2.170732259750366,
    "eval_logps/rejected": -3.2270374298095703,
    "eval_loss": 0.690127432346344,
    "eval_rewards/accuracies": 0.6913946866989136,
    "eval_rewards/chosen": -2.170732259750366,
    "eval_rewards/margins": 1.0563050508499146,
    "eval_rewards/rejected": -3.2270374298095703,
    "eval_runtime": 43.2304,
    "eval_samples": 1345,
    "eval_samples_per_second": 31.112,
    "eval_sft_loss": 2.133072853088379,
    "eval_steps_per_second": 7.795,
    "total_flos": 0.0,
    "train_loss": 0.6914210148065283,
    "train_runtime": 33926.4655,
    "train_samples": 59790,
    "train_samples_per_second": 5.287,
    "train_steps_per_second": 0.165
}